Data Science Wire

A Contextual-Bandit Oversight Game with Two-Sided Informational Asymmetry

arXiv cs.AI1mo4 min read

arXiv:2607.00155v1 Announce Type: new Abstract: We study runtime human oversight of an AI agent when private information runs in both directions: the human privately knows her reward function, while the AI privately knows the quality of the action it proposes. This is the kind of asymmetry that arises naturally when an autonomous robot or software agent has inspected a situation its human supervisor cannot directly assess. Building on Cooperative Inverse Reinforcement Learning (CIRL) and the Oversight Game, we introduce a contextual-bandit team game with two-sided asymmetric information and a

Read the full story at arXiv cs.AI

More in AI